基于知识图谱的建筑安全领域计算机视觉研究综述

建筑业作为高风险行业,建筑安全一直是行业关注的焦点。近年来,计算机视觉技术的快速发展为建筑工程安全管理提供了可能性。为全面、系统、客观地反映建筑安全领域计算机视觉技术的研究成果及其不足,运用知识图谱分析工具VOSviewer对来自WOS数据库核心合集上的166篇相关研究文献进行了可视化定量分析并辅以定性分析,分析了计算机视觉技术在建筑安全领域的研究现状与发展趋势。结果表明:计算机视觉技术在建筑安全领域应用的研究文献发表数量呈现上升趋势,并在近五年表现突出;该领域的作者研究合作群体分布分散,彼此缺乏合作;该领域的研究主题大致分为三个方向,即设备的识别与跟踪、工人不安全行为的识别以及计算机视觉及其发展;该领域未来的研究方向可能会尝试将计算机视觉技术与多种数字技术和深度学习技术相结合、增强个人防护设备检测的及时性,同时整体场景信息的获取以及综合安全监控系统设计相关技术的开发运用值得期待。

Overview of Computer Vision Research in Construction Safety Field Based on Knowledge Graph

As a high-risk industry,construction safety has always been the focus of the industry.In recent years,the rapid development of computer vision technology has provided huge possibilities for safety ma-nagement in the construction field.To comprehensively,systematically and objectively reflect the research results and research deficiencies of computer vision in the field of construction safety,this paper performs a visual quantitative analysis and qualitative analysis of 166 journal articles from the core collection of the WoS database,using the Knowledge Graph analysis tool,VOSviewer,and analyzes the research status and development trend of computer vision research in construction safety field.The results indicate that researches on computer vision in the field of construction safety maintain the upward trend,and has performed prominently in the past five years.The author’s cooperation groups of authors in this field are scattered and lack cooperation.The current research mainly focuses on three aspects,including equipment identification and tracking,worker’s unsafe behavior identification and computer vision and its development.Accordingly,future research may try to combine computer vision with multiple digital technologies and deep lear-ning,and enhance the timeliness of personal protective equipment detection.Meanwhile,the development and application of related technologies of overall scene information acquisition and comprehensive safety monitoring system design are worth looking forward to.